conf – conf.c:mount_entry:2019 – No such device or address

lxc容器重启时出现问题，提示：

Name: pang
Remote: unix://
Architecture: x86_64
Created: 2022/11/02 06:47 UTC
Status: Stopped
Type: persistent
Profiles: default

Log:

lxc pangyindong 20221105051756.396 ERROR    conf - conf.c:mount_entry:2019 - No such device or address - Failed to mount "/var/lib/lxd/devices/pangyindong/unix.gpu0.dev-nvidia-caps" on "/usr/lib/x86_64-linux-gnu/lxc/dev/nvidia-caps"
lxc pangyindong 20221105051756.396 ERROR    conf - conf.c:lxc_setup:3611 - Failed to setup mount entries
lxc pangyindong 20221105051756.396 ERROR    start - start.c:do_start:1263 - Failed to setup container "pangyindong"
lxc pangyindong 20221105051756.396 ERROR    sync - sync.c:__sync_wait:62 - An error occurred in another process (expected sequence number 5)
lxc pangyindong 20221105051756.396 WARN     network - network.c:lxc_delete_network_priv:2589 - Operation not permitted - Failed to remove interface "eth0" with index 18
lxc pangyindong 20221105051756.396 ERROR    lxccontainer - lxccontainer.c:wait_on_daemonized_start:842 - Received container state "ABORTING" instead of "RUNNING"
lxc pangyindong 20221105051756.396 ERROR    start - start.c:__lxc_start:1939 - Failed to spawn container "pangyindong"
lxc 20221105051756.397 WARN     commands - commands.c:lxc_cmd_rsp_recv:132 - Connection reset by peer - Failed to receive response for command "get_state"

原因是容器启动时试图挂载unix.gpu0.dev-nvidia-caps，但是这个是无法挂载的，所以出错。

我们可以ls一下/dev/nvidia*看看：

/dev/nvidia0  /dev/nvidia1  /dev/nvidiactl  /dev/nvidia-modeset  /dev/nvidia-uvm  /dev/nvidia-uvm-tools

/dev/nvidia-caps:
nvidia-cap1  nvidia-cap2

目前的解决方案是手动移除/dev/nvidia-caps这个文件夹。当然，安全起见，把这个文件夹换个名字吧。